Skip to main content

Visualizing Uncertain Weather

Storm prediction is tricky business. So is illustrating it

Credit:

National Weather Service

This article was published in Scientific American’s former blog network and reflects the views of the author, not necessarily those of Scientific American


Last week’s Winter Storm Stella demonstrated the fuzzy edges, shifting nature, and uncertainty inherent to storm forecasting. Early predictions had New York City prepping for a record-breaking—well, compared to past storms in March, anyway—20+ inches of snowfall. But Stella delivered less than eight inches of slush in Central Park. That said, the numbers weren’t too surprising for folks watching meteorologists hedge their bets with talk of shifting snowfall bands along the coast 24 hours before the precipitation began. (Even as the Weather Prediction Center cautiously held to their maximum predictions for NYC). And the storm lived up to the hype in places such as Binghamton, NY, with a 24-hour record-breaking snowfall of 31.3 inches

As Yoni Appelbaum (senior editor at The Atlantic) put it on Twitter:


On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.


Interestingly, though, in the days following the storm I didn’t see folks critiquing visualizations of precipitation uncertainty related to Stella. And when I think of Nor'easters, I think of daily updates of fairly standard radar images and maps with bands of precipitation projections, often in the form of low-to-high ranges. Hurricane projections, on the other hand, are nearly always accompanied by problematic “cone of uncertainty” visualizations. 

Credit: National Weather Service

As Kenneth Broad, Anthony Leiserowitz, Jessica Weinkle, and Marissa Steketee discuss in their article “Misinterpretations of the ‘cone of uncertainty’ in Florida during the 2004 hurricane season,” these ubiquitous maps are not designed to be the sole piece of information to inform personal risk and evacuation decisions. “Nonetheless, secondary evidence gleaned from other sources, including surveys of hurricane behavior, media reports, and in-depth interviews with hurricane forecasters, suggests that many members of the public pay close attention to this graphic, consider it as part of their decision-making process, and, at times, misinterpret it” (my emphasis).

And no wonder. Take another look at the map (below). How do you interpret the cone?

Credit: National Weather Service

As defined by the National Hurricane Center, the cone, “represents the probable track of the center of a tropical cyclone, and is formed by enclosing the area swept out by a set of circles (not shown) along the forecast track (at 12, 24, 36 hours, etc). The size of each circle is set so that two-thirds of historical official forecast errors over a 5-year sample fall within the circle.”

But as a visualization—even with the on-image text disclaimer “NOTE: The cone contains the probable path of the storm center but does not show the size of the storm. Hazardous conditions can occur outside of the cone.”— that full context isn’t terribly clear. 

Several weeks ago, Alberto Cairo, Mark Hansen, and I moderated a visualizing uncertainty workshop at the 2017 CAR Conference in Jacksonville, Florida (sponsored by Investigative Reporters and Editor’s National Institute for Computer Assisted Reporting program, with our session co-sponsored by the Society for News Design). As part of the introduction, Cairo showed this series of slides, based on a hypothetical hurricane:

What you show 
Credit: Alberto Cairo

What the cone is based on
Credit: Alberto Cairo

What I think some people see
Credit: Alberto Cairo

What non-scientists are not aware of (cone is just 66% probability)
Credit: Alberto Cairo

What we could be showing instead
Credit: Alberto Cairo

What this all may mean!
Credit: Alberto Cairo

Then, we asked the workshop participants to read critiques of the current cone of uncertainty design, including the 2004 article referenced above, “Perceptions of hurricane hazards in the mid-Atlantic region: Mid-Atlantic hurricane hazard perception,” By Michelle Saunders and Jason Senkbeil, and the academic poster “The power of maps to (mis)communicate: A case study of forecaster’s versus the public's interpretation of hurricane track maps” by Gina Eosco, and propose alternate design solutions suitable for a phone app. Here are some of the resulting concept sketches, borne out of brainstorming in groups.

Credit: Jen Christiansen (photograph)

Redesign exercises can be problematic, as they often completely ignore the very real constraints of the original design team, such as set data input formats, deadlines, budgets, compatibility and accessibility concerns, interests of other agencies, and beyond. And in a workshop context, we’re free to redefine the ultimate context and intended audience on the fly, to suit our fancy. But our goal wasn’t really to develop a viable and imminently useable alternative design to the cone of uncertainty. (Although wouldn’t it be fun if colored pencils really were the final output medium for a hurricane projection reboot). Our goal was to get visual journalists thinking deeply and critically about how uncertainty can be represented in graphical form. 

In that spirit, I watched Stella precipitation projections evolve in the days before the storm. And I couldn’t help but wonder—is the hurricane cone of uncertainty simply trying to accomplish too much? There’s something refreshingly simple about these experimental snow accumulation map sets from the National Weather Service, with their clear and candid titles:

Expect at Least This Much
Credit: National Weather Service

Potential for This Much
Credit: National Weather Service

After all, as a resident of the northeast, I wasn’t particularly interested in Stella’s exact path overhead. I was interested in best and worst case scenarios on the ground. The map pair above delivered that information to me clearly and succinctly.

I’m thrilled to see the National Weather Service experiment with new ways of communicating uncertain winter storm metrics, like predicted snow accumulations. Here’s hoping that they also have a new hurricane tracker up their sleeve.

Jen Christiansen is author of the book Building Science Graphics: An Illustrated Guide to Communicating Science through Diagrams and Visualizations (CRC Press) and senior graphics editor at Scientific American, where she art directs and produces illustrated explanatory diagrams and data visualizations. In 1996 she began her publishing career in New York City at Scientific American. Subsequently she moved to Washington, D.C., to join the staff of National Geographic (first as an assistant art director–researcher hybrid and then as a designer), spent four years as a freelance science communicator and returned to Scientific American in 2007. Christiansen presents and writes on topics ranging from reconciling her love for art and science to her quest to learn more about the pulsar chart on the cover of Joy Division's album Unknown Pleasures. She holds a graduate certificate in science communication from the University of California, Santa Cruz, and a B.A. in geology and studio art from Smith College. Follow Christiansen on X (formerly Twitter) @ChristiansenJen

More by Jen Christiansen